From: Tom Hughes Date: Wed, 26 Oct 2022 21:21:36 +0000 (+0100) Subject: Attempt to make juniper alarms more robust X-Git-Url: https://git.openstreetmap.org./chef.git/commitdiff_plain/d0b40e08ff15dd1253f6140ebbba8660be08d13b Attempt to make juniper alarms more robust --- diff --git a/cookbooks/prometheus/templates/default/alert_rules.yml.erb b/cookbooks/prometheus/templates/default/alert_rules.yml.erb index 417641cf0..df1329147 100644 --- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb +++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb @@ -217,12 +217,12 @@ groups: - name: juniper rules: - alert: juniper fan alarm - expr: jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!="running"} > 0 + expr: sum_over_time(jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!="running"}[5m]) > 0 for: 5m labels: alertgroup: "{{ $labels.site }}" - alert: juniper power alarm - expr: jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!="running"} > 0 + expr: sum_over_time(jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!="running"}[5m]) > 0 for: 5m labels: alertgroup: "{{ $labels.site }}"